i am having a little bit of trouble understanind what the Root Mean Squared Error (RMSE) or Root Mean Squared Error Deviation (RMSD), which i know are two names for the same thing, tell me about the quality of an estimation method.
i am starting to read simulation studies to inform applied researchers about how to choose the best statistical methods and many times i read that the RMSE is used to assess 'quality' or 'accuracy'. reading about it on wikipedia tells me that it works a little bit like a 'standard deviation' of sorts but i cannot make much sense of it after that. so if say i have two methods competing to be which one is the 'best'... do i prefer the one with the smallest RMSE? why? the wikipedia article says that "For an unbiased estimator, the RMSD is the square root of the variance, known as the standard error"
so, in this case, i should not necessarily look for the smallest RMSE but the one closes to the true standard errors? is that how it works?
i am a little bit confused and hope someone will be able to help
Advertise on Talk Stats